The most mature applications of 2022
Executive Summary
Logistics and manufacturing leaders are bombarded with abstract visions for how A.I. and other leading-edge technologies might one day improve their business.
Although the theoretical possibilities really are immense, the number of machine learning solutions that can be integrated into the daily operations of a typical facility and have a tangible impact and predictable R.O.I. is still small.
Identifying these mature services demands time-intensive research, and this is not an industry with an excess of spare time for its executives. Therefore, this article aims to offer a quick overview of some of the machine learning applications that are ‘ready for work’ today.
1. Verteego: Stock management and price automation for retail
Who’s using it already?
- Two of France’s largest consumer goods retail chains: Monoprix and Systeme U.
- Europcar, a rental firm that employs 9,000 people and turns over $3 billion a year.
- Moleskine, an Italian luxury stationery brand with about $160 million/year in revenue.
- A few e-commerce and omnichannel manufacturers, mostly headquartered in Europe.
Why should you pay attention?
Of course, stock-out forecasting is nothing new, and for established mass retailers there are plenty of industry standard solutions. But three of France’s biggest companies have decided that this startup offers something unique.
Part of the reason may be the potential for machine learning to bring the next competitive advantage in stock management. But the meat of the solution is the way it combines stock intelligence with pricing.
This is similar to the kind of technology Chinese omnichannel companies use to squeeze profit from each market fluctuation. At its most advanced, this approach sees sales and manufacturing exchanging data constantly, responding to changes in each other’s operations minute by minute.
Verteego might give established retailers in America and Europe a way to start developing such capabilities without interrupting their normal operations.
2. Fleetio: Fleet management with a lot of analytics
Who’s using it already?
- The American Automobile Association in Oregon and Idaho.
- Boyle Transportation, a high-security logistics provider HQ’d in Massachusetts.
- Delivery Mates, a whitelabelled last-mile service based in the UK.
- A number of complex fleet operators (airlines, public sector, construction firms etc).
Why should you pay attention?
Fleet management technology has advanced rapidly in the last few years. Rising and unpredictable fuel costs amongst other issues have made this an area of competitive advantage, even for businesses with relatively simple fleet operations.
There are mature solutions available from large, legacy vendors. But they are nowhere near Fleetio in terms of user experience and support. More important than that is the amount of data that the Fleetio platform can produce, and the simplicity of integrating it with other business systems.
If you manage your fleet with excel spreadsheets today, Fleetio might be the perfect foothold into digital maturity - and once you’re ready, it can grow with you, helping you get ahead with preventative maintenance and intelligently forecasting your fleet operating costs.
3. Tulip: Extremely Customizable Manufacturing Execution System
Who’s using it already?
- Electronics manufacturers, including Jabil ($30 billion / year).
- A few energy firms, including Europe’s Schneider Electric (also $30 billion / year).
- Some consumer goods brands, including New Balance ($4.4 billion / year)
- Merck, along with numerous smaller pharmaceutical and medical device manufacturers.
Why should you pay attention?
Tulip lays the foundation that will one day let manufacturers build a ‘digital twin’ of their entire production system.
The promise of the digital twin idea is that decision-makers can try out different operational strategies like they’re playing a game based on the real manufacturing environment. The problem with existing platforms is that there’s a big disconnect between what happens in the game and what happens in real life.
To bridge that gap we have to collect copious amounts of data from actual operations and feed them to an A.I. that gradually gets good at predicting what will happen. Tulip makes it easier than ever to collect that data, and already has the bones of that A.I.
Sounds like a leading-edge product best left to manufacturers with their own in-house software team right? Except Tulip is actually also a very intuitive MES with a drag-and-drop dashboard builder, so you can start getting actionable insights within weeks of integrating it.
4. Transfix: Smart Freight Brokerage with integrated TMS/FMS
Who’s using it already?
- Food and beverage firms like The Honey Baked Ham Company (est. $500 million / year)
- Packaging providers including Plastic Ingenuity (est. $133 million / year)
- Disruptive direct-to-consumer startups like FLOYD. (Doubled revenue 2019-2020)
- 3PLs and carriers of various sizes and specialties.
Why should you pay attention?
Transfix is certainly not the first or the biggest automated freight marketplace, nor the only freight brokerage with TMS and FMS functionality built in for shippers and carriers.
But it is the solution that combines a clean user interface, seamless integration with other systems, and serious machine intelligence under the hood into one package.
Its biggest competitors are pursuing a strategy of growth through acquisition. Ultimately Transfix might become part of one of those behemoths as well - and if it does, the resulting product will probably become the most advanced overnight.
As for today, transfix offers the majority of the functionality of one of the major players at a fraction of the cost, and presented in a way that even the most old school freight manager can learn within days. It’s the perfect service for shippers or carriers that know they need to catch up on digital maturity and want to do it fast.
5. DataDocks: Deep optimization for loading dock operations
Who’s using it already?
- S&P 100 firms including PepsiCo
- A few disruptive subscription services like HelloFresh and Stitch Fix.
- Food and beverage companies that use 3rd party carriers alongside an owned fleet.
- Many 3PLs and specialist transportation and warehousing providers.
Why should you pay attention?
Of course we can’t resist including ourselves on this list, but there’s a good reason for it.
Dock scheduling is a feature in many warehouse management systems and some transportation platforms as well. The problem is it’s never very good. Most shipping and receiving coordinators end up resorting to a plethora of Excel spreadsheets and spend their days on the phone with carriers and suppliers to confirm details of each load, and update the data manually.
DataDocks remedies that. It provides carriers with an effortless way to request appointments at your facility, and lets your coordinator drag-and-drop loads between times, days, docks, or the yard outside, instantly triggering notifications to everyone who needs to know about the change.
In terms of supply chain visibility, it builds a bridge between transportation and warehousing. Instead of disappearing when a truck is loaded or unloaded, data can now follow goods through their whole journey.
All of this is made possible by continuous improvement, and that’s where A.I. comes in. Over time, DataDocks learns the patterns of your loading dock operations - which carriers are usually an hour late, which days loads get worked the fastest, where the bottlenecks are - until your shopping and receiving process truly is a value center.
Further Reading
There is still a gap between how machine learning in supply chains is seen by academics, by software companies, and by real world practitioners.
But over time that gap will narrow. Reading some of the research papers published over the last few years, you’d think we were on the cusp of a sci-fi industrial revolution. It might seem far-fetched, but what the software featured in this article proves is that sometimes, the main difference between theory and practice is execution.
Here’s a few scientific papers that inspired us to find out which companies are leading the way in developing mature machine learning applications for supply chains:
- 2016: Predicting Future Inbound Logistics Processes using Machine Learning
- 2020: A Review of Further Directions for Artificial Intelligence, Machine Learning, and Deep Learning in Smart Logistics
- 2021: A systematic review of machine learning in logistics and supply chain management: current trends and future directions
